1 Resveratrol suppresses epithelial-to-mesenchymal transition in 2 colorectal cancer through TGF-β1/Smads signaling pathway 3 mediated Snail/E-cadherin expression 4 Qing Ji1,2,5, Xuan Liu1,5, Zhifen Han1, Lihong Zhou1, Hua Sui1, Linlin Yan1, Haili 5 Jiang1, Jianlin Ren3, Jianfeng Cai4, Qi Li1,* 6 7 Qing Ji1,2,5 8 E-mail: ttt99118@hotmail.com 9 Xuan Liu1,5 10 E-mail: lxtw@hotmail.com 11 Zhifen Han1 12 E-mail: zhifenghan1984@163.com 13 Lihong Zhou1 14 E-mail: zlhtcm@hotmail.com 15 Hua Sui1 16 E-mail: lqsh_2007@126.com 17 Linlin Yan1 18 E-mail: linlinyan2014@163.com 19 Haili Jiang1 20 E-mail: hailijiang2014@126.com 21 Jianlin Ren3 22 E-mail: jianlinren2014@163.com 23 Jianfeng Cai4 24 E-mail: jianfeng_cai67@163.com 25 Qi Li1,*: 26 E-mail: Lzwf@hotmail.com 27 28 1Department 29 Chinese Medicine, Shanghai 201203, China of Medical Oncology, Shuguang Hospital, Shanghai University of Traditional 1 30 2 31 Traditional Chinese Medicine, Shanghai 201203, China 32 3 33 Shanghai 200071, China 34 35 4Department 36 5 37 * 38 Shanghai University of Traditional Chinese Medicine, Shanghai 201203, China 39 E-mail address: Lzwf@hotmail.com 40 Abstract 41 Background: Resveratrol extracted from grape has been an ideal alternative drug in 42 the therapy of different cancers including colorectal cancer (CRC). Since the 43 underlying mechanisms of resveratrol on the invasion and metastasis of CRC have not 44 been fully elucidated, and epithelial-to-mesenchymal transition (EMT) is a key 45 process associated with the progression of CRC, here we aimed to investigate the 46 potential mechanism of resveratrol on the inhibition of TGF-β1-induced EMT in CRC 47 LoVo cells. 48 Methods: We investigated the anticancer effect of resveratrol against LoVo cells in 49 vitro and in vivo. In vivo, the impact of resveratrol on invasion and metastasis was 50 investigated by mice tail vein injection model and mice orthotopic transplantation 51 tumor model. In vivo imaging was applied to observe the lungs metastases, and 52 hemaoxylin-eosin (HE) staining was used to evaluate metastatic lesions from lung 53 metastases or hepatic metastases. In vitro, impact of resveratrol on the migration and 54 invasion of LoVo cells was evaluated by transwell assay. Inhibition effect of 55 resveratrol on TGF-β-induced EMT was examined by morphological observation. 56 Epithelial phenotype marker E-cadherin and mesenchymal phenotype marker 57 Vimentin were detected by western blot and immunofluorescence. The promoter 58 activity of E-cadherin was measured using a dual-luciferase assay kit. The mRNA 59 expression of Snail and E-cadherin was measured by RT-PCR. Research Center for Traditional Chinese Medicine and Systems Biology, Shanghai University of Department of Oncology, Shanghai Municipal Hospital of Traditional Chinese Medicine, of Chemistry, University of South Florida, Tampa, FL 33620, USA The authors contributed equally to this work. Corresponding author: Qi Li, M.D., Ph.D., Department of Medical Oncology, Shuguang Hospital, 2 60 Results: We demonstrated that, resveratrol inhibited the lung metastases of CRC 61 LoVo cells in vivo. In addition, resveratrol reduced the rate of lung metastases and 62 hepatic metastases in mice orthotopic transplantation. In vitro, TGF-β1-induced EMT 63 promoted the invasion and metastasis of CRC, reduced the expression of E-cadherin 64 and elevated the expression of Vimentin, and activated the TGF-β1/Smads signaling 65 pathway. But resveratrol could inhibit the invasive and migratory ability of LoVo 66 cells in a concentration-dependent manner, increased the expression of E-cadherin, 67 repressed the expression of Vimentin, as well as the inhibition of TGF-β1/Smads 68 signaling pathway. Meanwhile, resveratrol reduced the level of EMT-inducing 69 transcription factors Snail and the transcription of E-cadherin during the initiation of 70 TGF-β1-induced EMT. 71 Conclusions: Our new findings provided evidence that, resveratrol could inhibit EMT 72 in CRC through TGF-β1/Smads signaling pathway mediated Snail/E-cadherin 73 expression, and this might the potential mechanism of resveratrol on the inhibition of 74 invasion and metastases in CRC. 75 Keywords: Resveratrol; Colorectal cancer; Epithelial-to-Mesenchymal Transition; 76 TGF-β1/Smads signaling pathway; Snail; E-cadherin 77 Background 78 Colorectal cancer (CRC) is one of the leading causes of cancer-associated death in the 79 worldwide [1]. Many patients diagnosis of CRC are at advanced stages, and the 80 prognosis of these patients remains very poor. Generally, when the oncogenes 81 associated with cell proliferation were up-regulated, or the tumor suppressor genes 82 were down-regulated, the tumor cells would evade immune system and form tumors 83 in distal locations/organs through the pathway of invasion and metastases [2, 3]. 84 Epithelial-to-Mesenchymal Transition (EMT), a biological process occurs in various 85 types of epithelial cancers including CRC, is associated largely with increased 86 invasion and metastases [4-6]. EMT mainly experiences the following steps: 87 dissociation of adhesions between epithelial cells, loss of the apical-basolateral 3 88 polarity, reorganization of the actin cytoskeleton, and increases of cell motility. 89 Recently, numerous studies demonstrated that, some cytokines like TGF-β, HGF and 90 EGF can induce the process of EMT [7-9]. Meanwhile, various signaling pathways 91 associated with EMT were activated, such as TGF-β/Smads signaling pathway we 92 focused on. 93 Traditional Chinese Medicine (TCM), whether the formula or the extracted monomer, 94 have been identified as effective anticancer drugs in various cancers. Long-term basic 95 research and clinical application suggested that, resveratrol has been an ideal 96 alternative drug in the therapy of different diseases. Recently, anti-cancer activity of 97 resveratrol has been explored in various types of cancer including pancreatic cancer, 98 myeloma, ovarian cancer, breast cancer via the regulation of multiple pathways 99 [10-13]. Several experimental studies have demonstrated that, resveratrol played an 100 effective anti-cancer activity in the treatment of colorectal cancer including our own 101 completed research [14-16]. 102 However, the underlying molecular mechanisms through which resveratrol inhibits 103 migration and invasion of CRC cells have not been fully elucidated, and since EMT is 104 a key process associated with the progression of CRC, herein we aimed to investigate 105 the potential mechanism of resveratrol on the inhibition of TGF-β1-induced EMT in 106 CRC cells. 107 Methods 108 Materials 109 Recombinant human TGF-β1 was purchased from R&D. Resveratorl was purchased 110 from Sigma-Aldrich and dissolved at a concentration of 100 mM in DMSO as a stock 111 solution. Rabbit monoclonal antibodies against human E-cadherin, Vimentin, Slug, 112 Snail, ZEB1, Twist1, MMP-2, MMP-9, β-actin were purchased from Cell Signaling 113 Technology. Matrigel was purchased from BD Biosciences, and 24-well transwells 4 114 was purchased from Corning. 115 Cell culture 116 Human colorectal cancer cell line LoVo (ATCC, USA) was maintained in F12K 117 medium containing 10 % fetal bovine serum (FBS), 100 U/mL penicillin, 100 mg/mL 118 streptomycin, and incubated in a humidified, 5 % CO2 atmosphere, at 37 °C. 119 CCK assay for cell proliferation 120 Cell Counting Kit-8 (CCK-8) was selected to determine cell proliferation. Briefly, 121 LoVo cells were seeded in 96-well plates at 1×104 cells/well, when the cells reached 60 % 122 confluence, the medium was replaced with fresh medium containing different 123 concentrations of resveratrol, and incubated for 48 h and 72 h. The medium was then 124 discarded, and the cells were incubated with medium containing CCK-8 reagent for 4 125 hours. The absorbance was determined at 450 nm using a microplate reader (Biorad, 126 USA). All the experiments were repeated three times. 127 In vivo imaging by tail vein injection 128 Experimental lung metastases were achieved by injections of a single-cell suspension 129 of LoVo cells containing green fluorescent protein (GFP) into the lateral tail vein. One 130 week later, the mice were randomized into four groups of 8 animals each. Resveratrol 131 with a dose of 0, 50, 100, 150 mg/kg [17] was interfered in distinct groups via 132 intragastric administration every day for 3 weeks. Seven weeks later, prior to in vivo 133 imaging, the mice were anesthetized with penobarbital sodium, and the images of 134 established lung metastases were observed by LB983 NIGHTOWL II system 135 (Berthold Technologies GmbH, Germany). Afterwards, both of the lung organs were 136 excised, fixed with 10% neutral buffered formalin, and paraffin-embedded. The lung 137 sections were fully cut, and each section was set to 6 μm. All the lung sections were 138 stained with hemaoxylin-eosin (HE), following by counting the number of lung 139 metastases, and assessing comprehensively the extent of metastasis. All experimental 5 140 protocols were reviewed and approved by the animal ethics committee of Shuguang 141 Hospital, Shanghai University of Traditional Chinese Medicine. 142 Anti-tumor effect of resveratrol on mice with orthotopic transplantation tumor 143 Single-cell suspensions of LoVo cells (2 × 106 in 100 µL) were injected into the 144 subcutaneous area of female BALB/c nude mice (4-6 weeks old) obtained from SLAC 145 (SLAC Laboratory Lab, Shanghai, China). After the tumors reached the size of 100 146 mm3, the tumors were excised, fractionated, and transplanted into the appendix of the 147 nude mice. After two weeks, the mice were randomized into four groups of 8 animals 148 each. Resveratrol with a dose of 0, 50, 100, 150 mg/kg [17] was interfered in distinct 149 groups via intragastric administration every day for 3 weeks. After 42 days, animals 150 were sacrificed by cervical dislocation in deep CO2 anesthesia, primary tumors were 151 surgically removed and weighted (g). Then, a part of the removed tumors, the lung 152 and liver of the mice were investigated by HE staining. All experimental protocols 153 were reviewed and approved by the animal ethics committee of Shuguang Hospital, 154 Shanghai University of Traditional Chinese Medicine. 155 Western blot 156 Whole cell proteins were prepared according to the instructions of ProteoJET 157 Cytoplasmic Kit (Fermentas, USA). The extracted protein was quantified by BCA 158 protein assay. Proteins were loaded onto the SDS-PAGE gels for electrophoresis, 159 transferred to PVDF membranes, blocked in 5% milk, and incubated with the primary 160 antibodies following by the HRP-conjugated secondary antibodies. All the resulting 161 immunocomplexes 162 experiment was repeated independently three times. 163 Immunofluorescence microscopy 164 LoVo cells (2.5×105) were fixed for 40 minutes with 4% paraformaldehyde in PBS at 165 room temperature, blocked with 5% non-fat dry milk, and permealized with solution were visualized by enhanced 6 chemiluminescence. Each 166 containing 1% BSA, 0.5% Triton X-100. The cells were first stained with E-cadherin 167 rabbit antibody followed by Cy3-conjugated goat anti-rabbit IgG or first stained with 168 the Vimentin rabbit antibody followed by FITC-conjugated goat anti-rabbit IgG. 169 DAPI was applied for nuclear staining. Immunofluorescence images were taken with 170 a DMI3000B inverted microscope (Leica, Germany). All the experiments were 171 repeated three times. 172 Transwell assay for migration and invasion 173 LoVo cells (5×105, in F12K medium with 0.5% FBS) pretreated with or without 174 different concentration of resveratrol were seeded into the upper part of a transwell 175 chamber. For migration analysis, 600 µl F12K medium with 10 µg/ml fibronectin and 176 15% FBS was added in the lower part of the chamber, and the assay was performed. 177 Migrated cells were analyzed by crystal violet staining, followed by observing under a 178 DMI3000 B inverted microscope (Leica, Germany). Five random views were selected 179 to count the migrated cells. For invasion analysis, 100 µl matrigel (BD, USA) was 180 firstly added onto the bottom of the upper transwell chamber before LoVo cells were 181 seeded, and the following procedures were as same as migration analysis, except that 182 the invasive cells were analyzed after co-culture for 48 hours. Each experiment was 183 repeated independently three times. 184 RT-PCR 185 For 186 5-CAATCGGAAGCCTAACTA-3, 5-CAGATGAGCATTGGCAGCG-3, with control 187 GAPDH: 188 and the product were confirmed with agarose electrophoresis. All assays were 189 performed in triplicate and independently repeated three times. 190 Plasmid constructions RT-PCR of Snail gene, the primers 5-GAAGGCTGGGGCTCATTTG-3, 7 were designed as follows: 5-GGGCCATCCACAGTCTTC-3, 191 Human Snail gene (Gene ID: 6615) was amplified by RT-PCR from LoVo cells, using 192 the 193 5-CGGGATCCTCAGCGGGGACATCC-3, sequenced by the Sangon Biotech 194 company (Shanghai, China), and the right Snail fragments (795 bp) were sub-cloned 195 into the expression vector of pcDNA3.1, named pcDNA3.1-Snail. 196 Analysis of the E-cadherin promoter 197 To test E-cadherin promoter activity, LoVo cells were co-transfected with either the 198 recombinant plasmid pGL3-basic-E-cadherin or pGL3-basic-mut-E-cadherin with a 199 control positive plasmid pRL-SV40. The promoter activity was measured using a 200 dual-luciferase assay kit (Beyotime Institute of Biotechnology, China) according to 201 the manufacturer’s instructions. 202 Statistical analysis 203 All the data were presented as mean ± standard deviation ( X ± S) and analyzed with 204 SPSS18 Software. The mean values of two groups were compared by Student’s t test. 205 P < 0.05 was considered statistically significant, and P < 0.01 was considered 206 significant difference. 207 Results 208 Resveratrol inhibited the metastases of colorectal cancer LoVo cells in vitro and 209 in vivo imaging by tail vein injection 210 First, we determined the cytotoxic effect of resveratrol on colorectal cancer LoVo 211 cells using CCK assay. As shown in Fig. 1A, LoVo cells were treated with various 212 concentrations of resveratrol (0, 6.25, 12.5, 25, 50, 100, 200 μM) for 48 h and 72 h. It 213 was observed that, resveratrol inhibited the proliferation of LoVo cells in a 214 concentration- and time-dependent manner. After 48 h of resveratrol (12.5 μM) 215 treatment, cell viability was reduced by approximately 10 %, and this data indicated forward and reverse primers: 5-CCGCTCGAGATGCCGCGCTCTTT-3, 8 216 that less than 12.5 μM resveratrol had little influence on the cell proliferation of LoVo 217 cells, while more than 50 μM resveratrol significantly inhibited the proliferation of 218 LoVo cells (P < 0.01). 219 To observe the effect of resveratrol on LoVo cells in vivo, we performed the 220 experiment of in vivo imaging by tail vein injection. From the results we found that, 221 LoVo cells could partly migrate to the lung organs of the mice after tail vein injection 222 for seven weeks, and resveratrol of different concentration (0, 50, 100, 150 mg/Kg) 223 could inhibit the metastatic ability of LoVo cells in a concentration-dependent manner 224 (Fig. 1B). However, resveratrol has no effect on the body weight of the mice (data not 225 shown). Then, the organs of lung were excised, and the metastatic lesions were 226 determined by HE staining. From the imaging results we found, a lot of metastatic 227 lesions were seen in the lung organs with no treatment of resveratrol. Nevertheless, 228 when the increased resveratrol was treated, the metastatic lesions decreased gradually, 229 especially at 150 mg/Kg, few metastatic lesions were found (Fig. 1C). 230 Resveratrol inhibited the invasive ability of the orthotopic transplantation tumor 231 To investigate thoroughly the invasive ability of the orthotopic transplantation tumor 232 originated from colorectal cancer LoVo cells, we made a long term experiment in vivo. 233 As shown in the final results, the metastatic lesions were found in the district of liver 234 and lung organs in the mice without resveratrol treatment (Fig. 2A). But, with the 235 increased treatment of resveratrol, the metastatic lesions in the liver and lung organs 236 decreased. At 150 mg/Kg resveratrol, the metastatic lesions were rarely found (Fig. 237 2B). Additionally, with the increase of resveratrol from 50 mg/Kg to 150 mg/Kg, the 238 finally excised tumors decreased in a dose-dependent manner (Fig. 2C), while 239 resveratrol had no effect on the body weight of the mice (data not shown). 240 Resveratrol inhibited the morphological changes of TGF-β1-induced EMT 241 Since resveratrol could inhibit the metastases and invasion of LoVo cells, we have 9 242 enough reason to interpret the effect mechanism of resveratrol. We sought to 243 determine whether resveratrol could inhibit TGF-β1-induced EMT in LoVo cells. The 244 images demonstrated that, LoVo cells showed a mesenchymal phenotype after 245 treatment with 10 ng/mL TGF-β1 for 48 h, but in the LoVo cells treated with 10 246 ng/mL TGF-β1 and resveratrol of 6 and 12 μM simultaneously, the mesenchymal 247 phenotype showed fewer gradually (Fig. 3A). For the characterization of EMT, we 248 detected E-cadherin and Vimentin by western blot and immunofluorescence. From the 249 western blot, we found, after adding TGF-β1 to induce EMT for 48 h, the expression 250 of E-cadherin decreased obviouslywhile the Vimentin increased remarkably, and 251 resveratrol could reverse the shifted expression of E-cadherin and Vimentin induced 252 by TGF-β1 (Fig. 3B). From the imaging of immunofluorescence we knew, after 253 adding TGF-β1 to induce EMT, the expression of E-cadherin in the membrane 254 decreased, but the expression of Vimentin in the cytoplasm increased obviously. But 255 when resveratrol was added, the expression of E-cadherin and Vimentin approached 256 gradually to the previous level without addition of TGF-β1. All above findings 257 indicated that resveratrol could inhibit the effects of TGF-β1 on EMT in LoVo cells 258 (Fig. 3C). 259 Resveratrol inhibited TGF-β1-induced migration and invasion in colorectal 260 cancer LoVo cells 261 When the TGF-β1 induced EMT, not only the morphology of LoVo cells changed, 262 but also their biological functions were different. By the transwell assay, we found 263 that, being treated with 10 ng/mL TGF-β1, the migratory and invasive ability of LoVo 264 cells increased obviously. However, resveratrol could inhibit the migration and 265 invasion of LoVo cells caused by the TGF-β1 induced EMT (Fig. 4A). Accordingly, 266 we observed the factors of MMP2 and MMP9, which were associated closely with 267 invasion and migration. The results demonstrated that, the expressions of MMP2 and 268 MMP9 were consistent with the shifted ability of invasion and migration of LoVo 269 cells (Fig. 4B). 10 270 Resveratrol repressed expressions of EMT-induced transcription factors and 271 Smads signaling pathway 272 To examine the inhitory ability of resveratrol on the expression of EMT-induced 273 trascription factors, the expression of Slug, Snail, ZEB1 and Twist were detected by 274 western blot. The results showed that, in contrast to the control LoVo cells, Snail 275 up-regulated significantly in LoVo cells treated with 10 ng/ml TGF-β1. Nevertheless, 276 resveratrol 277 concentration-dependent manner (Fig. 5A). 278 Considering that the TGFβ/Smad signaling pathway is a critical pathway triggered by 279 phosphorylation of Smads, we consequently measured the active status of Smads 280 signaling pathway. Not surprisingly, we found the expression of p-Smad2/3 elevated 281 in TGF-β1 treated LoVo cells, and the Smad2/3 level decreased accordingly. When 282 resveratrol was added simultaneously, the levels of p-Smad2/3 and Smad2/3 turn back 283 to the former levels without any treatment (Fig. 5B). 284 To confirm the effect of resveratrol on inhibiting the TGFβ/Smad signaling pathway, 285 we used the TGFβRI/II inhibitor (LY2109761) to inhibit the TGFβ/Smad signaling 286 pathway and observed the downstream effectors. From the RT-PCR results we knew 287 that, when the TGFβ1 was added, the mRNA expression of Snail increased, but the 288 E-cadherin decreased. When the TGFβRI/II inhibitor was added, the mRNA 289 expression of Snail and E-cadherin gradually returned back to former levels, 290 respectively (Fig. 5C). Accordingly, the protein expression of Snail and E-cadherin 291 was consistent with their mRNA expression. Additionally, the TGFβRI/II inhibitor 292 could also inhibit the phosphorylation of Smad2/3 induced by TGFβ1 in LoVo cells 293 (Fig. 5D). The results implied that, resveratrol may regulate the EMT-associated 294 protein expression through the TGFβ/Smad signaling pathway. To be a supplementary 295 explanation, we also found that resveratrol had little impact on the expression of 296 TGFβ1 or TGFβRI/II (data not shown). could inhibit the Snail 11 level induced by TGF-β1 in a 297 Resveratrol suppressed E-cadherin expression via inhibiting the transcription of 298 Snail and reducing its binding on the promoter of E-cadherin 299 In order to understand the detailed mechanism of resveratrol on EMT, we examined 300 the effect of resveratrol on the mRNA expression of Snail and E-cadherin. As shown 301 in Fig. 6A, resveratrol could inhibit the transcription of Snail and promote the 302 transcription of E-cadherin induced by TGFβ1 in a concentration-dependent manner. 303 When the Snail was overexpressed in LoVo cells, the mRNA levels of E-cadherin 304 were down-regulated obviously (Fig. 6B). Accordingly, the protein expression of 305 Snail and E-cadherin changed consistently with their mRNA levels. This indicated 306 that Snail could affect the transcription of E-cadherin (Fig. 6C). For confirming above 307 results, we measured the effect of resveratrol and Snail on the promoter of E-cadherin. 308 It demonstrated that, whether the resveratrol or the overexpressed Snail, they could 309 promote the transcription activity of E-cadherin (Fig. 6D and E). 310 Discussions 311 Various plant or fruit-derived agents with few side effects have been accepted as 312 potential alternatives for the therapy of colorectal cancer. Resveratrol extracted from 313 grape or Polygonum cuspidatum is a natural antioxidant, which can reduce blood 314 viscosity, maintain the blood flow, and inhibit the platelet aggregation [18-22]. In 315 addition, resveratrol has anti-cancer activity in a great number of malignant tumors 316 like prostate, skin, ovarian, breast and colon cancers [10-13, 16]. 317 Our previous studies have revealed the potential therapeutic effect of resveratrol 318 against invasion and metastasis of colorectal cancer cells [16]. In that study, we found 319 resveratrol inhibited invasion and metastasis through MALAT1 mediated β-catenin 320 signaling pathway. Here, we observed a previously unknown mechanism, in which 321 resveratrol 322 Epithelial-to-Mesenchymal Transition induced by TGFβ1. could inhibit invasion 12 and migration via reversing 323 EMT is characterized by the loss of cell-cell adhesion and the increase of cell motility, 324 and it is a key process in cancer progression and metastasis [5, 6], which making the 325 inhibition of EMT process an attractive therapeutic strategy. EMT could be triggered 326 by many growth factors like TGF-β, EGF, HGF [7-9, 13, 2223]. Our present studies 327 demonstrated that TGF-β1-induced LoVo cells undergo morphological alterations 328 characteristic of EMT characterized by up-regulated expression of mesenchymal 329 markers Vimentin and down-regulated expression of E-cadherin epithelial markers 330 including and increased metastasis and invasion,. up-regulated expression of 331 mesenchymal markers Vimentin and down-regulated expression of E-cadherin 332 epithelial markers. TGF-β1 also enhances enhanced the expression of zinc-finger 333 transcriptional factors Snail, which then repressed the E-cadherin transcription. These 334 transcriptional repressors of E-cadherin are required during EMT development [5]. 335 The results of thisOur study showed that resveratrol reduced migration and invasion 336 in a concentration-dependent manner and inhibited TGF-β1-induced EMT in LoVo 337 cells, as proved by the increase of the expression of E-cadherin and the decrease of 338 Vimentin. During EMT development, TGF-β induced the Snail expression, and the 339 increased Snail transcriptional factor would inhibit the promoter activity of 340 E-cadherin, leading to the decreased E-cadherin transcription. In addition, our results 341 showed that, the expression of EMT inducing transcription factors Snail could be 342 inhibited by resveratrol effectively. 343 Although EMT is a coordinated, organized program involving interaction between 344 different cells and tissue types, the EMT program could be activated in response to 345 alterations of microenvironment, which would contribute to occurrence of the 346 diseases including cancer progression [24, 25]. We observed that, TGF-β1-induced 347 EMT promoted the invasion and migration ability of LoVo cells, but reseveratrol 348 could inhibit the promoting effect in a concentration-dependent manner. Expression 349 analysis also demonstrated that treatment of resveratrol significantly down-regulated 350 MMP2 and MMP9 induced by TGF-β1. Therefore, resveratrol might inhibit the 351 invasion and metastasis of CRC cells by suppressing TGF-β1-induced EMT. 13 352 TGF-β/Smad signaling pathway is a classical pathway associated closely with the 353 proliferation, differentiation, migration, and so on. In this system, TGF-β1 regulates 354 cellular 355 (TGF-βRI/TGF-βRII), and the activated TGF-βRI phosphorylates Smad2 or Smad3, 356 will bind to Smad4 [26, 27]. The resulting Smad complex then moves into the nucleus, 357 where it interacts in a cell-specific manner with various transcription factors to 358 regulate the transcription of many genes [23, 28]. Snail was one of TGF-β/Smad 359 signaling pathway mediated gene [29, 30], which repressed the E-cadherin expression, 360 promoted the EMT process, and finally increased the ability of invasion and 361 metastasis of CRC cells in our study. However, resveratrol could inhibit the invasion 362 and metastasis by preventing the continuation of EMT process. 363 Conclusions 364 In summary, our study provided evidence that resveratrol could inhibit EMT in 365 colorectal 366 Snail/E-cadherin expression, and this might the potential mechanism of resveratrol on 367 inhibition of invasion and metastases (Fig. 7). 368 Abbreviations 369 CRC, 370 hemaoxylin-eosin; TCM, Traditional Chinese Medicine; CCK-8, Cell Counting Kit-8; 371 GFP, green fluorescent protein; FBS, fetal bovine serum. 372 Competing interests 373 The authors declare that they have no competing interests. 374 Authors’ contributions 375 Conceived and designed the experiments: QL QJ JR CJ. Performed the experiments: 376 QJ XL ZH LY HJ. Analyzed the data: QJ XL ZH HS HZ JR JC. Contributed processes cancer Colorectal by binding through cancer; and phosphorylating TGF-β1/Smads EMT, signaling cell-surface pathway Epithelial-to-mesenchymal 14 receptors mediated transition; HE, 377 reagents/materials/analysis tools: ZD JZ HM FJH. Wrote the paper: QJ XL QL JR JC. 378 All authors have read and approved the final manuscript. 379 Acknowledgements 380 This work was supported by National Natural Science Foundation of China 381 (81303102, 81473478, 81303103, 81473628), Program of Shanghai Municipal 382 Education Commission (12YZ058), Shanghai Municipal Health Bureau (2011ZJ030, 383 20114037), Chen Guang project of Shanghai Municipal Education Commission and 384 Shanghai Education Development Foundation (13CG47). 385 References 386 1. 387 Watson AJ, Collins PD: Colon cancer: a civilization disorder. Dig Dis 2011, 29(2):222-228. 388 2. Christofori G: New signals from the invasive front. Nature 2006, 441(7092):444-450. 389 3. Fidler IJ: The pathogenesis of cancer metastasis: the ‘seed and soil’ hypothesis revisited. 390 391 Nat Rev Cancer 2003, 3(6):453-458. 4. 392 393 2002, 2(6):442-454. 5. 394 395 Thiery JP, Acloque H, Huang RY, Nieto MA: Epithelial-mesenchymal transitions in development and disease. Cell 2009, 139(5):871-890. 6. 396 397 Thiery JP: Epithelial-mesenchymal transitions in tumour progression. Nat Rev Cancer Guarion M: Epithelial-mesenchymal transition and tumour invasion. Int J Biochem Cell Biol 2007, 39(12):2153-2160. 7. Wendt MK, Schiemann WP: Therapeutic targeting of the focal adhesion complex 398 prevents oncogenic TGF-beta signaling and metastasis. Breast Cancer Res 2009, 399 11(5):R68. 400 8. 401 402 403 Tian M, Neil JR, Schiemann WP: Transforming growth factor-beta and the hallmarks of cancer. Cell Signal 2011, 23(6):951-962. 9. Tirino V, Camerlingo R, Bifulco K, Irollo E, Montella R, Paino F, Sessa G, Carriero MV, Normanno N, Rocco G: TGF-beta1 exposure induces epithelial to mesenchymal 15 404 transition both in CSCs and non-CSCs of the A549 cell line, leading to an increase of 405 migration ability in the CD133+ A549 cell fraction. Cell Death Disease 2013, 4:e620. 406 10. Shankar S, Nall D, Tang SN, Meeker D, Passarini J, Sharma J, Srivastava RK: Resveratrol 407 inhibits pancreatic cancer stem cell characteristics in human and KrasG12D transgenic 408 mice by inhibiting pluripotency maintaining factors and epithelial-mesenchymal 409 transition. PLoS One 2011, 6(1):e16530. 410 11. Sun C, Hu Y, Liu X, Wu T, Wang Y, He W, Wei W: Resveratrol downregulates the 411 constitutional activation of nuclear factor-B in multiple myeloma cells, leading to 412 suppression of proliferation and invasion, arrest of cell cycle and induction of apoptosis. 413 Cancer Genet Cytogenet 2006, 165(1):9-19. 414 12. Vergara D, Simeone P, Toraldo D, Del Boccio P, Vergaro V, Leporatti S, Pieragostino D, 415 Tinelli A, De Domenico S, Alberti S, Urbani A, Salzet M, Santino A, Maffia M: Resveratrol 416 downregulates Akt/GSK and ERK signaling pathways in OVCAR-3 ovarian cancer 417 cells. Mol Biosyst 2012, 8(4):1078-1087. 418 13. Vergara D, Valente CM, Tinelli A, Siciliano C, Lorusso V, Acierno R, Giovinazzo G, 419 Santino A, Storelli C, Maffia M: Resveratrol inhibits the epidermal growth 420 factor-induced epithelial mesenchymal transition in MCF-7 cells. Cancer Lett 2011, 421 310(1):1-8. 422 14. Mahyar-Roemer M, Katsen A, Mestres P, Roemer K: Resveratrol induces colon tumor cell 423 apoptosis independently of p53 and precede by epithelial differentiation, mitochondrial 424 proliferation and membrane potential collapse. Int J Cancer 2001, 94(5):615-622. 425 15. Delmas D, Rebe C, Lacour S, Filomenko R, Athias A, Gambert P, Cherkaoui-Malki M, 426 Jannin B, Dubrez-Daloz L, Latruffe N, Solary E: Resveratrol-induced apoptosis is 427 associated with Fas redistribution in the rafts and the formation of a death-inducing 428 signaling complex in colon cancer cells. J Biol Chem 2003, 278(42):41482-41490. 429 16. Ji Q, Liu X, Fu XL, Zhang L, Sui H, Zhou LH, Sun J, Cai JF, Qin JM, Ren JL, Li Q: 430 Resveratrol Inhibits Invasion and Metastasis of Colorectal Cancer Cells via MALAT1 431 Mediated Wnt/β-Catenin Signal Pathway. PLoS One 2013, 8(11):e78700. 432 17. Majumdar AP, Banerjee S, Nautiyal J, Patel BB, Patel V, Du J, Yu Y, Elliott AA, Levi E, 433 Sarkar FH: Curcumin synergizes with resveratrol to inhibit colon cancer. Nutr Cancer 16 434 435 436 437 2009, 61(4):544-53. 18. Baur JA, Sinclair DA: Therapeutic potential of resveratrol: the in vivo evidence. Nat Rev Drug Discov 2006, 5(6):493-506. 19. Tomé-Carneiro J, Gonzálvez M, Larrosa M, Yáñez-Gascón MJ, García-Almagro FJ, 438 Ruiz-Ros JA, Tomás-Barberán FA, García-Conesa MT, Espín JC: 439 Grape resveratrol increases serum adiponectin and downregulates inflammatory genes 440 in peripheral blood mononuclear cells: a triple-blind, placebo-controlled, one-year 441 clinical trial in patients with stable coronary artery disease. Cardiovasc Drugs Ther 2013, 442 27(1):37-48. 443 20. Koz ST, Etem EO, Baydas G, Yuce H, Ozercan HI, Kuloğlu T, Koz S, Etem A, Demir N: 444 Effects of resveratrol on blood homocysteine level, on homocysteine induced oxidative 445 stress, apoptosis and cognitive dysfunctions in rats. Brain Res 2012, 1484:29-38. 446 21. Carrizzo A, Forte M, Damato A, Trimarco V, Salzano F, Bartolo M, Maciag A, Puca 447 AA, Vecchione C: Antioxidant effects of resveratrol in cardiovascular, cerebral and 448 metabolic diseases. Food Chem Toxicol 2013, 61:215-226. 449 22. Wightman EL, Reay JL, Haskell CF, Williamson G, Dew TP, Kennedy DO: Effects 450 of resveratrol alone or in combination with piperine on cerebral blood flow parameters 451 and cognitive performance in human subjects: a randomised, double-blind, 452 placebo-controlled, cross-over investigation. Br J Nutr 2014, 112(2):203-213. 453 23. Geng J, Fan J, Ouyang Q, Zhang X, Zhang X, Yu J, Xu Z, Li Q, Yao X, Liu X, Zheng J: 454 Loss of PPM1A expression enhances invasion and the epithelial-to-mesenchymal 455 transition in bladder cancer by activating the TGF-β/Smad signaling pathway. 456 Oncotarget 2014, 5(14):5700-5711. 457 24. Jung HY, Fattet L, Yang J: Molecular Pathways: Linking Tumor Microenvironment to 458 Epithelial-Mesenchymal 459 doi:10.1158/1078-0432.CCR-13-3173. 460 461 462 463 Transition in Metastasis. Clin Cancer Res 2014, 25. O'Connor JW, Gomez EW: Biomechanics of TGF-β-induced epithelial-mesenchymal transition: implications for fibrosis and cancer. Clin Transl Med 2014, 3:23. 26. Derynck R, Zhang YE: Smad-dependent and Smad-independent pathways in TGFbeta family signalling. Nature 2003, 425(6958):577-584. 17 464 465 27. Blobe GC, Schiemann WP, Lodish HF: Role of transforming growth factor beta in human disease. N Engl J Med 2000, 342(18):1350-1358. 466 28. Peng ZX, Fan QM, Du L, Yan W, Tang TT, Tu B: Osteosarcoma cells promote the 467 production of pro-tumor cytokines in mesenchymal stem cells by inhibiting their 468 osteogenic differentiation through the TGF-β/Smad2/3 pathway. Exp Cell Res 2014, 469 320(1):164-173. 470 29. Wen SL, Gao JH, Yang WJ, Lu YY, Tong H, Huang ZY: Celecoxib attenuates hepatic 471 cirrhosis through inhibition of epithelial-to-mesenchymal transition of hepatocytes. J 472 Gastroenterol Hepatol 2014, doi: 10.1111/jgh.12641. 473 30. Li Y, Sun Y, Liu F, Sun L, Li J, Duan S, Liu H, Peng Y, Xiao L, Liu Y, Xi Y, You Y, Li H, 474 Wang M, Wang S, Hou T: Norcantharidin inhibits renal interstitial fibrosis by blocking 475 the tubular epithelial-mesenchymal transition. PLoS One 2013, 8(6):e66356. 476 477 Legends of figures 478 479 480 481 482 483 484 485 486 Figure 1 Resveratrol inhibited the metastasis of colorectal cancer LoVo cells in vitro and in vivo imaging by tail vein injection. (A) Correlation of resveratrol drug concnetrations (0, 6.25, 12.5, 25, 50, 100, 200 μM) and cell viability in LoVo cells for 48 h and 72 h. (B) LoVo-pLV4-GFP cells were respectively injected into the lateral tail vein. One week later, resveratrol was administrated with the concentration of 0, 50 mg/Kg, 100mg/Kg, 150 mg/Kg every day for 3 weeks. Seven weeks later, the established lungs metastases images were observed by LB983 NIGHTOWL II system. (C) The organs of lung were excised, and the metastasis was checked by hemaoxylin-eosin staining, and the numbers of metastatic lesions were counted. **P < 0.01, compared with LoVo-pLV4-GFP cells without treatment of resveratrol. 487 488 489 490 491 492 493 494 Figure 2 Resveratrol inhibited the invasive ability of the orthotopic transplantation tumor. (A) Resveratrol with a dose of 0, 50, 100, 150 mg/kg was interfered in distinct groups for the orthotopic transplantation tumor mice via intragastric administration every day for 3 weeks. After 42 days, HE staining was performed to check the numbers of metastatic lesions. (B) The numbers of the lung metastases and hepatic metastases were counted, **P < 0.01, compared with group without treatment of resveratrol. (C) The finally excised primary tumors were weighed, **P < 0.01, compared with group without treatment of resveratrol. 495 496 Figure 3 Resveratrol inhibited the morphological changes of TGF-β1-induced EMT. (A) 18 497 498 499 500 501 502 503 504 LoVo cells were treated with 10 ng/ml TGF-β1 to induce EMT, and resveratrol with a concentration of 6, 12 μM were introduced to inhibit the morphological changes. Control LoVo cells displayed classical epithelial morphology, and 10 ng/ml TGF-β1 treated LoVo cells represented a mesenchymal phenotype. (B) Expression of epithelial phenotype marker E-cadherin and mesenchymal phenotype marker Vimentin were detected by western blot, **P < 0.01, compared with control LoVo cells without treatment of TGF-β1 and resveratrol. (C) Immunofluorescence staining of E-cadherin and Vimentin in LoVo cells treated with 10 ng/ml TGF-β1, with or without resveratrol. 505 506 507 508 509 510 511 512 Figure 4 Resveratrol inhibited TGF-β1-induced migration and invasion in colorectal cancer LoVo cells. (A) LoVo cells treated with 10 ng/ml TGF-β1, with or without resveratrol (6, 12 μM) were applied for the trasnwell experiment. Migration and invasion of indicated LoVo cells were quantified. Values represent the number of migratory/invasive cells per 5 high power fields. **P < 0.01, compared with group without treatment of TGF-β1 and resveratrol. (B) MMP2 and MMP9 were determined by western blot. *P < 0.05, **P < 0.01, compared with group without treatment of TGF-β1 and resveratrol. 513 514 515 516 517 518 519 520 521 522 Figure 5 Resveratrol repressed expressions of EMT-induced transcription factors and Smads signaling pathway. Whole or cell extracts of different cellular compartments from LoVo cells were probed for (A) Slug Snail, ZEB1, Twist, and (B) p-Smad2/3, Smad2/3. **P < 0.01, compared with group without treatment of TGF-β1 and resveratrol. (C) LoVo cells treated with TGF-β1 or (and) TGFβRI/II inhibitor (LY2109761) were analyzed for the Snail and E-cadherin mRNA expression by RT-PCR, GAPDH was chosen as a control. (D) Whole or cell extracts of different cellular compartments from LoVo cells were probed for p-Smad2/3, Smad2/3, Snail and E-cadherin expression. **P < 0.01, compared with group without treatment of TGF-β1 and LY2109761. 523 524 525 526 527 528 529 530 531 532 533 534 535 536 Figure 6 Resveratrol suppressed E-cadherin expression via inhibiting the transcription of Snail and reducing its binding on the promoter of E-cadherin. (A) LoVo cells treated with TGF-β1 or (and) resveratrol were analyzed for the Snail and E-cadherin mRNA expression by RT-PCR. **P < 0.01, compared with group without treatment of TGF-β1 and resveratrol. (B) LoVo cells trasfected with pcDNA3.1-Snail or pcDNA3.1 plasmid were analyzed for the Snail and E-cadherin mRNA expression by RT-PCR. **P < 0.01, compared with the empty vector group. (C) Whole or cell extracts of different cellular compartments from LoVo cells expressing pcDNA3.1-Snail or pcDNA3.1 were probed for Snail and E-cadherin. **P < 0.01, compared with the empty vector group. (D) The relative E-cadherin promoter activities were detected in LoVo cells treated with TGF-β1 or (and) resveratrol. **P < 0.01, compared with group without treatment of TGF-β1 and resveratrol. (E) The relative E-cadherin promoter activities were detected in LoVo cells expressing pcDNA3.1-Snail or pcDNA3.1. **P < 0.01, compared with the empty vector group. 19 537 538 539 Fig. 7 A hypothetical illustration for the mechanism of resveratrol on the EMT of colorectal cancer. 20